EP0579922B1 - Data transmission device for a local area network - Google Patents

Description

The invention relates to a data transmission arrangement for a local data network which in the preamble of the claim 1 Art.

Such data transmission arrangements are advantageous in a local data network, which is connected to a asynchronous data transfer protocol that works out individual UART characters (bit serial character) is constructed.

It is generally known for the transmission of binary signals apply the AMI coding (alternate mark inversion).

An AMI coding device is known (DE 37 35 395 C2), where an arithmetic unit is used for coding becomes.

There is also a serial transmitter and receiver unit known (GB 22 38 931 A), which for coupling a network node to a transmission cable of a data network Transformer with a center tap used. this shows also a terminal unit (EP-A-0 316 536), which for Coupling a transmitter and receiver unit to one Transmission cable a transformer with a center tap used. In addition, this terminal unit has one Receiving and a transmission unit on the convert received or to be transmitted signals and from or read or write to the data network.

From FR-A-2 567 702 is for the control of the for Coupling the relevant transformer a constant current source known with two symmetrical current mirrors.

The invention is based, for a local task Data network with a changeable tree-shaped topology to create a reliable data transmission arrangement, which is the construction of an inexpensive network driver unit for enables a network node.

The invention consists in the specified in claim 1 Characteristics. Advantageous configurations result from the dependent claims.

Fig. 1

Teile eines Netzwerk-Knotens mit einer Netz-Treiber-Einheit für ein Übertragungsmedium,

Fig. 2

ein Datennetzwerk mit mehreren Netzwerk-Knoten und einem Netzwerk-Abschluß,

Fig. 3

ein Schaltbild der Netz-Treiber-Einheit,

Fig. 4

die Sprungantwort eines Signals in einem Kabelstrang des Datennetzwerkes,

Fig. 5

den prinzipiellen Verlauf von Signalen der Netz-Treiber-Einheit nach der Fig. 3 bei der Decodierung und

Fig. 6

den Verlauf der Signale bei der Codierung.

Show it:

Fig. 1

Parts of a network node with a network driver unit for a transmission medium,

Fig. 2

a data network with several network nodes and a network termination,

Fig. 3

a circuit diagram of the network driver unit,

Fig. 4

the step response of a signal in a cable harness of the data network,

Fig. 5

the basic course of signals of the network driver unit according to FIG. 3 in the decoding and

Fig. 6

the course of the signals during coding.

In Fig. 1, 1 means a network driver unit in which a code generator 2, a decoding unit 3 and a coupling element 4 are arranged for a transmission medium 5. The network driver unit 1 and a serial transmission / reception unit 6 are parts of a network node 7.

The transmit / receive unit 6 has an output 8 for bit serial Data TxD and an input 9 for bit serial data RxD and is advantageously a serial communication unit (serial communication interface SCI) of a microcomputer or a UART (universal asynchronous receiver / transmitter).

The code generator 2 has a coding unit 10 and one with the Coding unit 10 connected constant current stage 11 with a first output pole 12 for a current i and a second Output pole 13 for the current i. An input 14 of the coding unit 10 is with the output 8 of the transmission / reception unit 6 connected.

The decoding unit 3 has one with the input 9 of the Transmitter / receiver unit connected output 15 and two Input poles 16 and 17, of which a first input pole 16 with the first output pole 12 of the constant current stage 11 and one first input pole 18 of the coupling element 4 is connected, while the second input pole 17 to the second output pole 13 the constant current stage 11 and to a second input pole 19 of the coupling element 4 is connected.

The transmission medium 5 consists of at least one cable harness 5.1, which has at least two cores, one of which a first wire 20 with a first connection pole 22 of the Coupling element 4 and also a first connection 23 of a Network termination 24 and a second core 21 with a second Connection pole 25 of the coupling element 4 and a second Connection 26 of the network termination 24 is connected.

The network termination 24 is advantageously one of an ohmic Resistor 27 and a capacitor 28 existing series circuit. The capacitor 28 separates the resistor 27 direct current from the transmission medium 5.

In Fig. 2, seven network nodes 7 are local Data network 29 on the from a number s cable harnesses existing transmission medium 5 connected to a first wiring harness 5.1 has the network termination 24.

The network termination 24 is in principle at any point of the transmission medium 5 can be placed.

The number s is generally not limited and is in data network 29 shown four. On branch branches 30 (designated 30.1 to 30.4 in FIG. 2) are two in each case Cable harnesses interconnected in such a way that the topology of the data network 29 is in principle tree-shaped, the Tree branched in any way and therefore degenerate may.

One wire 20 or 21 of the first wiring harness 5.1 is on a first branch junction 30.1, each with a wire 20 or 21 a second cable run 5.2 and a second line branch 30.2 with one core 20 or 21 of a third Harness 5.3 connected. Furthermore, one wire is 20 each or 21 of the third wiring harness 5.3 on a third wiring branch 30.3 with one wire 20 or 21 of a fourth Wiring harness 5.4 connected.

In the data network 29, the maximum k is the number of connected Network nodes 7 are generally not restricted. In favor of an acceptable maximum response time for a token passing Process and also to get a small address range, the maximum k is advantageously limited to 127. The sum of the Lengths of all cable runs 5.1, 5.2, 5.3, 5.4 are in favor of one Another advantage, namely no line amplifiers (repeaters) to use, limited to about 1000 m. If necessary Data networks 29 for larger expansions in a known manner coupled via line amplifier.

An advantageous embodiment of the Network driver unit 1 with the coupling element 4 and bipolar constant current stage 11 has one over two Signal inputs and a first output (31) for a Signal Pos and a second output (32) for a signal Neg having comparator circuit (33), one of an OR gate 36 and a dual up / down counter 37 Signal filter 38, a polarity detector 39 with a Output 40 for a signal RP, one from a flip-flop 41, an exclusive-OR gate 42 and a first frequency divider 43 constructed generator part 44 and a quartz oscillator 45 with a second frequency divider connected downstream of the quartz oscillator 45 46 (prescaler) on.

The first input pole 16 is a first signal input Comparator circuit 33, while the second input pole 17 through the second signal input of the comparator circuit 33 is realized.

The comparator circuit 33 is connected to its first output 31 a first input 51 of the OR gate 36 and a first Control input 52 of the polarity detector 39 connected while the second output 32 to a second input 53 of the OR gate 36 and to a second control input 54 of the polarity detector 39 is connected.

A control input 55 for a control signal IN of the up / down counter 37 is with an output 56 of the OR gate 36 connected. The output 15 of the decoding unit 3 (FIG. 1) is on Output of the up / down counter 37 and at the same time also an output of the signal filter 38. The output 40 of the polarity detector 39 is connected to a data input 57 of the flip-flop 41, which one with advantage at a clock input 58 negative or falling edge triggerable D flip-flop (data Latch) with a data output 59, the data input 57 is the D input of the D flip-flop. The data output 59 for one Signal Pol of the flip-flop 41 has a first input 60 of the exclusive-OR gate 42, of which a second Input 61 with an output 62 for a signal TB of the first Frequency divider 43 is connected.

The bipolar constant current stage 11 has a first with a Output 63 of the exclusive-OR gate 42 connected control input 64 for a control signal TP and a second with the Input 14 of the network driver unit 1 connected control input 65 for the bit serial data TxD. In the further leads the input 14 to the clock input 58 of the multivibrator 41 and to a reset input 66 of the first frequency divider 43 and to a reset input 67 of the second frequency divider 46.

One intended for an advantageously binary clock signal CLK Output 68 of the second frequency divider 46 leads to a clock input 69 of the up / down counter 37, to a clock input 70 of the polarity detector 39 and to a clock input 71 of the first frequency divider 43.

The coupling element 4 advantageously has one with a first one Winding 72 and a second winding 73 provided Transformer 74 and a capacitor 75, the two Input poles 18 and 19 the connections of the first winding 72 and one between the two connection poles 22 and 25 the capacitor 75 and the second winding 73 existing Series connection lies.

The capacitor 75 decouples the transformer 74 in a direct current manner from the transmission medium 5.

The comparator circuit 33 is advantageously designed such that that common mode noise, which in essentially due to the capacity of transformer 74 occur, are effectively suppressed.

If necessary, between the two connection poles 22 and 25, parallel to that of capacitor 75 and the second Winding 73 existing series connection a voltage limiting element switched.

A higher layer asynchronous data transfer protocol (Upper layer) of the data transmission network 29 is off Protocol characters - for example from ASCII characters (American standard code for information interchange) - built which in the network node 7 between the transmitting / receiving unit (6) and the network driver unit (1) in the form of the bit serial data TxD or RxD can be exchanged.

The log characters of the bit serial data TxD or RxD have a constant length, that is, a number n of bits per The protocol character of the data transmission protocol is constant.

In addition, each protocol character of the bit serial data TxD or RxD has a number n ₁ start bits, a number n ₂ actual information bits, a number n ₃ check bits and a number n ₄ stop bits.

The number n ₁ and the number n ₂ , the number n ₃ and the number n ₄ advantageously correspond to a known standard, which also specifies a sequence of the individual bits of the protocol characters.

Each protocol character of the bit serial data TxD or RxD points advantageously an even number of bits with the value "0" or then an even number of bits with the value "1", so that with a low circuitry outlay from one log character a signal can be generated which over the duration of the Log character averaged has the value zero.

The n bits of the protocol character are in the network node 7 between the network driver unit (1) and the transmit / receive unit 6 and also between the network nodes 7 of the data network 29 with a stable bit rate f _bit , that is within the Protocol character transmitted bit-synchronously.

The code generator 2 shown in FIG. 1 transforms each Protocol characters of the bit serial data TxD in a ternary Code, which is advantageous as a in sections essentially constant current i in the two output poles 12 and 13 is shown. The over the duration of a log character average value of current i is advantageously zero.

The ternary code is advantageously a known AMI code (Alternate Mark Inversion), a first value of the code advantageously being a positive rectangular pulse of the current i with a pulse time T _bit , a second value being a negative rectangular pulse of the current i with the pulse time T _bit and the third value is shown with i = 0. The amount of current i is advantageously substantially constant during the pulse time T _bit . The pulse time T _bit results from the bit rate f _bit with the formula T _bit = 1 / f _bit .

The binary value that is always in an even number is used as an impulse in the AMI code transmitted, the polarity of a subsequent pulse is always reversed from its previous impulse.

The current i essentially flows through the input poles 18 and 19 of the coupling element 4, which transforms the pulses of the current i into corresponding pulses of a voltage u _22-25 lying between the two connection poles 22 and 25.

The constant current stage 11 has the advantages over a voltage stage that resonance phenomena of an oscillating circuit essentially formed from the transformed capacitance of the transmission medium 5 and the leakage inductance of the transformer 74 are effectively prevented and that the voltage u _22-25 on the transmission medium 5 increases with increasing distance from the network Termination 24 increased due to the line resistance of the transmission medium 5.

One acting between the two input poles 18 and 19 Input impedance of the coupling element 4 is advantageous at least about 1000 times smaller than that between the input poles 16 and 17 acting input impedance of the decoding unit 3.

The decoding unit 3 converts a voltage u _18-19 corresponding to the AMI-coded protocol character and lying between the two input poles 16 and 17 into the bit-serial data RxD of the protocol character.

The decoding unit 3 advantageously works simultaneously with the code generator 2 and in doing so transmits to the coding unit 10 the polarity of a previously registered pulse of the voltage u _18-19 .

One between the two wires 20 and 21 of the transmission medium 5 lying AC voltage is in the network node 7 through the coupling element 4 at its two input poles 18 and 19 to the input poles 16 and 17 of the decoding unit 3 transfer.

The coupling element 4 essentially fulfills four tasks, namely the transmission of the code generated by the code generator 2 in the form of a current i from the network node 7 to the transmission medium 5, the transmission of a code in the form of a voltage u _22-25 from the transmission medium 5 to Decoding unit 3 of the network node, direct current decoupling of the network node 7 from the transmission medium and electrical isolation of the network node 7 from the transmission medium 5 and thus electrical isolation of all network nodes 7 of the data network 29.

The essential parameters of the coupling element 4, the transmission medium 5 and the network termination 24 are matched to one another in a known manner at the desired bit rate f _bit . The fact that the code generator 2 generates the code in the form of the current i, which is essentially constant in sections, suppresses the resonance phenomena, the network topology (FIG. 2) of the data network 29 and the number of network nodes 7 in the data network 29 are described in those Limits can be changed and the voltage u _22-25 on the transmission medium 5 to a rectangular pulse of the current i always has an advantageous step response.

4 shows a positive pulse of the length T _{bit of} the current i and the advantageous step response of the voltage u _22-25 to the positive edge of the pulse. Delayed by a time of at most 0.35 · T _bit on the positive edge, the voltage u _{22-25 runs} within a rectangular window 77. The voltage u _22-25 in window 77 deviates at most 15% from its nominal value at logic "1". The window 77 limits the guaranteed course of the advantageous step response.

In the following, the mode of operation of that in FIG. 3 illustrated advantageous embodiment of the network driver unit 1 described.

The log characters of the bit serial data are TxD and RxD for this version according to one in the standard DIN 19 245 (Process Field Bus, PROFIBus) use UART characters assuming which one in the order listed Start bit with the value "0", eight information bits, one check bit and has a stop bit with the value "1". The check bit is a Parity bit, which by its value is a sum of the bits, whose value is "1" sets to an even value, where the information bits and the check bit of the Are taken into account.

The network driver unit 1 is based on simple considerations for other protocol characters customizable.

An AMI code is encoded in the network driver unit 1 or decoded with the value "0" of the log character as a pulse is shown. The log character has an even number Bits with the value "0" and the bit serial data TxD or RxD point between two protocol characters (i.e. in the so-called Idle state) the value "1".

A basic course of signals in FIG. 5 explains, by way of example, essential steps in the decoding of the AMI code of the voltage u _18-19 .

A time scale that applies to all signals in FIG. 5 is shown in FIG. 5a. The time unit T corresponds to the pulse time T _bit and has a duration of 52µs, for example, with a bit rate f _bit of 19200 s ^-1 .

The frequency f _{CLK of} the clock signal CLK is a multiple of the bit rate f _bit . With a bit rate f _bit = 19200 s ^-1 , the clock signal CLK advantageously has the frequency f _CLK of 614.4 kHz.

The binary signal Pos (FIG. 5b) at the first output 31 of the comparator circuit 33 is only "1" in the case of a positive pulse of the voltage u _18-19 , while the binary signal Neg (FIG. 5c) at the second output 32 of the comparator circuit 33 with a negative pulse of the voltage u _18-19 to logic "1".

The binary signal RP (FIG. 5e) at the output 40 of the polarity detector 39 represents the current polarity of the pulses of the voltage u _18-19 largely without interference; with logic "1" a positive pulse, with logic "0" a negative pulse. Advantageously, the polarity detector 39 is an up / down counter with a count Z ₃₉ (FIG. 5d), which can be changed between zero and an upper value b by means of pulses of the clock signal CLK. If the signal Pos is logic "1" and the counter Z _{39 is} less than the value b, a pulse of the clock signal CLK increases the counter Z ₃₉ by one if the signal Neg is logic "1" and the counter Z _{39 is} greater than zero is, a pulse of the clock signal CLK decreases the counter reading Z ₃₉ by one. If the counter reading Z _{39 reaches} the value b, the signal RP becomes logic "1" and retains the value until the counter reading Z ₃₉ reaches zero, whereupon the signal RP changes to logic "0".

The value b determines the interference immunity of the polarity detector 39. In an equation b / f _CLK = α · T _bit , the factor α is advantageously approximately 0.2 to 0.3. If the pulse time T _bit = 52µs and the frequency f _CLK = 614.4 kHz, then b = 7 is an advantageous value.

The signal IN (Fig. 5f) is formed from the two signals Pos and Neg, has the course (Pos OR Neg) and controls the counting direction of the up / down counter 37, the counter reading Z ₃₇ (Fig. 5g) by means of pulses Clock signal CLK is variable between zero and an upper value d. If the signal IN is logic "1" and the count Z _{37 is} less than the value d, a pulse of the clock signal CLK increases the count Z ₃₇ by one, if the signal IN is logic "0" and the count Z _{37 is} greater than zero is, a pulse of the clock signal CLK decreases the counter reading Z ₃₇ by one. If the counter reading Z _{37 reaches} the value d, the signal ¬RxD inverse to the bit serial data RxD (FIG. 5h) becomes logic "1" and retains the value until the counter reading Z ₃₉ reaches zero, whereupon the signal ¬RxD changes to logic " 0 "changes.

Since the signal IN is formed and used when decoding the AMI code, which in principle represents "pulse of the AMI code" and "no pulse of the AMI code", the positive and negative pulses of the voltage are u _18-19 when decoding basically equal, whereby the two connection poles 22 and 25 of the network driver unit 1 can be interchanged without loss of function, which is advantageous when connecting the network node 7.

The value d determines the effect of the signal filter 38. In an equation d / f _CLK = β · T _bit , the factor β is advantageously approximately 0.2 to 0.3, provided that the comparator circuit 33 has a small switching hysteresis (typically approximately 5% the threshold voltage). If the pulse time T _bit = 52µs and the frequency f _CLK = 614.4 kHz, then d = 7 is an advantageous value.

5i, a first arrow indicates the start bit of the bit serial data RxD and further arrows the times at which the data RxD sampled in the transmission / reception unit 6 will. The current information value for the arrows is bit serial data RxD specified.

A basic course of signals in FIG. 6 explains, by way of example, essential steps in the coding of the AMI code of the current i or the voltage u _18-19 (FIG. 6c) from the bit-serial data TxD (FIG. 6b).

A time scale that applies to all signals in FIG. 6 is shown in FIG. 6a. The time unit T corresponds to the pulse time T _bit and has a duration of 52µs, for example, with a bit rate f _bit of 19200 s ^-1 .

The binary signal RP (FIG. 6d or FIG. 5e) represents the polarity of the last determined pulse of the voltage u _18-19 , which determines the polarity of the subsequent pulse in the AMI code.

The fact that the coding depends on the polarity of the last determined pulse of the voltage u _18-19 ensures that successive pulses of the voltage u _18-19 always have different polarities, so that the transformer 74 is also driven symmetrically when the Network driver unit 1 alternates between receiving and sending.

Due to the flip-flop 41, the binary signal Pol (Fig. 6e) after a negative edge of the signal TxD the instantaneous value of the Signal RP at the time of the negative edge.

The first frequency divider 43 with a division q ₁ forms the signal TB from the clock signal CLK (FIG. 6f), the frequency f _{TB of which is given} by the formula f _{TB -f CLK} / q ₁ . The division is restarted by a negative edge at the reset input 66. The pulse time of the signal TB is advantageously the same as its pause time. The frequency f _TB is determined according to the relationship f _TB = f _bit / 2 by the bit rate f _bit .

For example, if the bit rate is f _bit = 19200 s ^-1 and the frequency f _TB = 614.4 kHz, then the division is q ₁ = 64.

The control signal TP (Fig. 6g) is from the two signals Pol and TB formed and has the course (Pol XOR TB), where XOR the means Boolean antivalence.

The polarity of the current i is in the constant current stage 11 determined by the control signal TP, which ensures that the polarity of subsequent pulses of the current i always changes.

The control signal TP also ensures that the polarity of a pulse of the current i is reversed from the polarity of a pulse of the voltage u _18-19 that preceded this current pulse. The constant current stage 11 selects the polarity of the current (i) based on the control signal TP such that successive pulses of the voltage (u _18-19 ) between the two input poles (18; 19) have a different polarity.

The current i in is transmitted via the second binary control input 65 the constant current stage 11 on or off, the Current switches on when TxD = "0" and switched off when TxD = "1" becomes.

The constant current stage 11 essentially consists, for example from a constant current source that can be switched off and from four to an H-bridge controlled controllable switches.

The second frequency divider 46 with a division q ₂ forms the clock signal CLK from the output signal of the quartz oscillator 45. The division is restarted by a negative edge at the reset input 67.

The cable strands of the transmission medium are advantageous conventional twisted pair cable.

The steepness of the step response of the voltage u _22-25 (FIG. 4) and thus the limits of the window 77 are determined by the leakage inductance L _{S of} the transformer 74 (FIG. 3) that can be determined when the second winding 73 is short-circuited, by the ohmic resistor 27 (FIG 1) of the network termination 24 and can be influenced by a supply voltage available from the constant current stage 11 (FIG. 3) for generating a pulse of the current i. In particular, the constant current stage 11 enables the advantageous increase in the step response in a simple manner.

At the supply voltage of 10 V, the advantageous step response results when the resistor 27 has a nominal value of 100 ohms and the leakage inductance L _{s is} between 0.75 mH and 2.25 mH.

The two windings 72 and 73 advantageously have the same Inductance L, which is advantageously greater than 1.5 H.

Due to the network termination 24, the transmission medium 5 is always advantageously completed, so that the transformer 74 especially when moving from sending to a pause due to its inductance, no interference pulses on the transmission medium 5 caused.

Because the transformer 74 through the capacitor 75 and the Resistor 27 of network termination 24 through capacitor 28 DC separated from the transmission medium 5 can the two wires 20 and 21 additionally transmit one Serve DC voltage with which the network node 7 with little effort can be fed via the transmission medium 5.

Through the data transmission arrangement with the two-wire, on anywhere with the network termination 24 completed transmission medium 5, with the network driver unit 1, which the decoding unit 3, the code generator 2 with the constant current stage 11 and the coupling element 4 with the Transformer 74, the topology of the data network 29 with simple means within wide limits small risk of malfunction. The network driver unit 1 is reliable and inexpensive to manufacture.

Claims (7)

1. A data transmission arrangement for a local data network (29) (local area network LAN) which operates with an asynchronous data transmission protocol which is made up from constant-length protocol characters represented in bit-serial mode and to which there belong at least two network nodes (7) each having a respective network driver unit (1) and with at least two respective connecting poles (22; 25) for a common transmission medium (5), wherein the transmission medium (5) is at least one cable strand (5.1) which has at least two wires (20;21) and wherein the network driver unit (1) has an encoding unit (10), a decoding unit (3) with an output (15) for the protocol characters and a coupling member (4) with two connecting poles (22; 25) for the transmission medium (5),
      characterised in that

   [a] the encoding unit (10) has an input (14) for the protocol characters,

   [b] the network driver unit (1) has a bipolar constant current stage (11) connected downstream of the encoding unit (10),

   [c] the coupling member (4) has two input poles (18; 19) wherein

   [c.1] the first input pole (18) of the coupling member (4) is connected to a first output pole (12) of the constant current stage (11) and a first input pole (16) of the decoding unit (3), and

   [c.2] the second input pole (19) of the coupling member (4) is connected to a second output pole (13) of the constant current stage (11) and a second input pole (17) of the decoding unit (3), and

   [d] the transmission medium (5) is terminated at a location by the two wires (20; 21) being connected at the location by a network termination (24).
2. A data transmission arrangement according to claim 1 characterised in that

   [a] a ternary code can be produced by the encoding unit (10) for each protocol character which is inputted at its input (14), the three values of which ternary code can be impressed into the input poles (18; 19) of the coupling member (4) by the bipolar constant stage (11) in respective portions substantially constantly in terms of magnitude as a positive pulse of a current (i) or as a negative pulse of the current (i) or as a current-less space, wherein the mean value of the current (i) is zero within a protocol character and wherein the polarity of one of the pulses of the current (i) is so selected that successive pulses of a voltage (u_18-19) between the two input poles (18; 19) are of different polarities,

   [b] the encoding unit (10) is connected to the decoding unit (3) in such a way that the direction of the current (i) can be communicated to the encoding unit (3), and

   [c] in the decoding unit (3) the positive pulses and the negative pulses are equated so that the two connecting poles (22; 25) of the coupling member (4) are interchangeable.
3. A data transmission arrangement according to one of claims 1 and 2 characterised in that the coupling member (4) has a transformer (74) with at least windings (72; 73) and a capacitor (75), wherein the two input poles (18; 19) are the connections of a first winding (72) and disposed between the two connecting poles (22; 25) is a series circuit comprising the capacitor (75) and a second winding (73).
4. A data transmission arrangement according to one of claims 1 to 3 characterised in that the network termination (24) is a series circuit comprising an ohmic resistor (27) and a capacitor (28).
5. A data transmission arrangement according to one of claims 1 to 4 characterised in that an AMI-code (alternate mark inversion) can be produced by the encoding unit (10) from the protocol characters.
6. A data transmission arrangement according to one of claims 1 to 5 characterised in that

   [a] the decoding unit (3) has

   [a.1] a comparator circuit (33) which has two signal inputs and a first output (31) and a second output (32), and

   [a.2] a signal filter (38) comprising an OR-gate (36) and a dual up/down counter (37), wherein

   [a.3] a first input pole (16) is the first signal input of the comparator circuit (33) and the second input pole (17) is the second signal input of the comparator circuit (33),

   [b] the encoding unit (10)

   [b.1] has a polarity detector (39) with an output (40) for a signal (RP),

   [b.2] a generator portion (44) comprising a trigger circuit (41), an exclusive-OR-gate (42) and a first frequency divider (43), and

   [b.3] a second frequency divider (46), wherein

   [b.4] the comparator circuit (33) is connected with its first output (31) to a first input (51) of the OR-gate (36) and a first control input (52) of the polarity detector (39), and wherein

   [b.5] the second output (32) is connected to a second input (53) of the OR-gate (36) and to a second control input (54) of the polarity detector (39),

   [c] a control input (55) for a control signal (IN) of the up/down counter (37) is connected to an output (56) of the OR-gate (36), [d] the output (15) of the decoding unit (3) is an output of the up/down counter (37) and at the same time also an output of the signal filter (38),

   [e] the output (40) of the polarity detector (39) is connected to a data input (57) of the trigger circuit (41) which has a clock input (58) and a data output (59) for a signal (Pol),

   [f] the data output (59) of the trigger circuit (41) is connected to a first input (60) of the exclusive-OR-gate (42) of which a second input (61) is connected to an output (62) for a signal (TB) of the first frequency divider (43),

   [g] the bipolar constant current stage (11) has a first control input (64) for a control signal (TP), which is connected to the output (63) of the exclusive-OR-gate (42), and a second control input (65) connected to the input (14) of the network driver unit (1),

   [h] the input (14) of the encoding unit (10) is additionally connected to the clock input (58) of the trigger circuit (41) and to a reset input (66) of the first frequency divider (43) and to a reset input (67) of the second frequency divider (46), and

   [i] the output (68) of the second frequency divider (46) is connected to a clock input (69) of the up/down counter (37), to a clock input (70) of the polarity detector (39) and to a clock input (71) of the first frequency divider (43).
7. A data transmission arrangement according to one of claims 1 to 6 characterised in that the polarity detector (39) is an up/down counter.

Images